The Untold Story of Google’s Quest to Bring the Internet Everywhere—By Balloon BY STEVEN LEVY08.13.136:30 AM On October 16, 2012, the granddaughter of one of Allen Epling’s houseguests spotted a shiny object in the sky. Epling, an amateur astronomer living in Pike County, Kentucky, grabbed his binoculars and spied a glimmering, tubelike shape hovering ominously high above. Along with his wife and guests, he watched it for more than two hours. “It wasn’t anything I recognized,” Epling later told a local reporter. Plenty of others saw the object too: The Kentucky State Police received multiple reports of sightings. A couple of days later, the Appalachian News-Express ran a story headlined “Mystery Object in Sky Captivates Locals.” Regional television stations reported that government agencies professed ignorance. The story was picked up by CNN. And the UFO-loving website Ashtar Command Crew linked to the news as ostensible proof of continuing visits from the Galactic Federation fleet. Epling, for his part, didn’t jump to extraterrestrial conclusions. But still, what was it? Rich DeVaul knows. Sitting in a conference room in Mountain View, California, he beams proudly as he runs a YouTube clip of one of the newscasts. The mysterious craft was his doing. Or, at least, the work of his Google team. The people in Pike County were witnessing a test of Project Loon, a breathtakingly ambitious plan to bring the Internet to a huge swath of as-yet-unconnected humanity—via thousands of solar-powered, high-pressure balloons floating some 60,000 feet above Earth. Google is obsessed with fixing the world’s broadband problem. High-speed Internet is the electricity of the 21st century, but much of the planet—even some of the United States—remains in the gaslit era; only about 2.7 billion earthlings are wired. Of course, it’s also in the company’s strategic interests to get more people online—inevitably, visitors to the web click on Google ads. Project Loon balloons would circle the globe in rings, connecting wirelessly to the Internet via a handful of ground stations, and pass signals to one another in a kind of daisy chain. Each would act as a wireless station for an area about 25 miles in diameter below it, using a variant of Wi-Fi to provide broadband to anyone with a Google-issued antenna. Voilà!—low-cost Internet to those who otherwise wouldn’t have it. The smartphones to connect to it are quickly becoming cheap. Over the years, Google has embarked on a number of pilot projects. In the US, it’s building its own high-speed networks in cities like Kansas City, Missouri, Austin, Texas, and Provo, Utah; it’s also lobbying to allocate unused slices of the television spectrum, called white spaces, for Internet access. But these approaches are too expensive or logistically daunting for much of the rest of the world that remains unwired. And so, Google’s quest to design a low-cost Internet service led it to a solution in a surprising place: the skies. On June 15, after two years of development, Google unveiled Project Loon at a press conference in Christchurch, New Zealand. Even as prime minister John Key spoke, a few of the 30 antenna-equipped balloons were still floating over the Pacific after having supplied a tiny and temporary bit of Internet access to some 50 local families. Google X chief Astro Teller (left) and Project Loon’s first leader, Rich DeVaul, holding the system’s ground-based antennas. Eric Ray Davidson Could this number expand to 50,000? 50 million? 500 million? Billions? That’s what Google hopes. Project Loon has the official status of a Google “moon shot,” a high-risk, high-reward, Hail Mary effort. Many of these moon shots involve computational challenges—like self-driving cars. But in this case, Google’s success hinges on its mastery of ballooning, a centuries-old craft whose mysteries have never been fully understood. The project will require steering these vehicles with more precision and for longer periods of time than anyone has ever managed. To pursue Project Loon, the company known for its algorithms has had to mindmeld with the world of Jules Verne’s aerial adventurer, Phileas Fogg. It’s a steampunk novel come to life. Google X is probably the only outfit in the world that would spend millions of dollars on Internet ballooning. Founded in early 2010, the research lab is specifically devoted to moon shots like self-driving cars and Google Glass. DeVaul, an MIT-trained engineer who wrote his doctoral thesis on “memory glasses,” arrived there in mid-2011 after a stint doing secret research at Apple. DeVaul joined X’s small Rapid Evaluation team, whose assignment was to triage concepts, mercilessly separating the so-crazy-they-just-might-work ideas from the just plain crazy ones. “They are mentally plastic in their ability to see the world differently,” says Astro Teller, who runs the X lab. Teller gave DeVaul some ideas to kill. One of them was a concept to deliver wireless Internet access via balloons in the stratosphere. CEO Larry Page had often spoken of this, and Teller knew that favorite topics of the cofounders had a leg up in funding decisions. As it turned out, technologists—including DeVaul—had been pondering the promise of balloon-based communication for years. But there was a big problem: Balloons are hostages to wind. If you try to keep a balloon in a fixed location, you must apply Sisyphean efforts to resist that wind. It almost always ends badly. Lockheed Martin recently tried to beat the odds with a giant solar-powered dirigible. But in its maiden test in 2011, Lockheed’s High Altitude Long Endurance-Demonstrator prototype failed to reach altitude and was forced to abort, landing in a Pennsylvania forest. Lockheed has no plans for another test. As DeVaul began spreadsheeting the possibilities, he came up with another concept. Rather than a behemoth that required massive amounts of energy to fight stratospheric winds to stay in place, he found himself drawn to the idea of smaller, cheaper weather balloons that sometimes stay aloft for 40 days or more, circling the globe. “I thought, why not have a bunch of these things, covering a whole area? How crazy would that be?” he says. There was an obvious flaw in DeVaul’s idea: It is incredibly difficult to steer balloons as they make their weeks-long journey around Earth. But what if the balloons could adjust their altitude in a way that took advantage of wind currents? You could maneuver them to rise or fall, allowing them to catch a ride in the desired direction. The key would be analyzing the voluminous data about wind currents, past and present, available from the US government’s National Oceanic and Atmospheric Administration. Crunching data just happened to be something DeVaul’s employer did very well. With the proper meteorological skill, an expertise in simulation, and a lot of computation, DeVaul surmised, you could have balloons traverse the globe in the same way sailboats use the shifting winds to arrive at the proper port. And like sailboats, they need no fuel for locomotion. The plan had kind of a hacker elegance, the equivalent of displacing an outdated mainframe computer with countless smaller servers, like the ones in Google’s data centers. Teller thinks this approach has implications far beyond computing and ballooning. “Once we see how much better, cheaper, and safer we can make things by adding intelligence,” he says, “all the things we think of now as solved will be thought of as being very much version 1.0. This is going to play out over and over again in our lifetimes.” But before DeVaul could build his fleet of balloons, he needed to ensure that balloon-borne Wi-Fi would work. One day in August 2011 he conducted his first test. He used a cheap “sounding” balloon with an envelope—the inflatable part—made of latex rubber, which isn’t designed to withstand high altitudes. (As a balloon goes higher, the pressure differential between the expanding trapped gas and the outside atmosphere increases; unless it has a very tough skin, the balloon will burst. Ultimately, Google would use high-pressure balloons with skins that could withstand those destructive differentials.) He had approached some Google engineers responsible for a PR stunt in which a Bugdroid figurine (the green Android mascot) soared miles over Earth. “If you can fly Android to the edge of space on a sounding balloon, maybe you could help me fly a little Linux computer with a Wi-Fi radio pointing downward,” DeVaul said to them. He called these the Icarus tests, named after a young character in Greek mythology whose wax wings melted when he flew too close to the sun. The group drove in DeVaul’s Subaru Forester to Dinosaur Point at the San Luis Reservoir in California’s Central Valley. They filled four latex balloons—bought online for about $100 each—with helium purchased from a welding supplier. Each balloon had a small Wi-Fi transmitter that could communicate with receivers on the ground. And then they let them go. The balloons rose steadily. And started moving east. Fast. Opening Photo: Trey Ratcliff
Posted on: Fri, 27 Sep 2013 20:53:05 +0000
Trending Topics
Recently Viewed Topics
© 2015